1
SAN JOSÉ STATE UNIVERSITY
College of Social Work
S. W. 242
Spring 2009
Edward Cohen
Week 9
March 20, 2009
•
Statistics review
•
Correlation—Pearson
r
•
Lab 2b Independent
t
-test
•
Final Paper discussion—Choosing a statistical procedure
•
Next week: Happy Spring Break!
Concepts you should know:
•
Review univariate, bivariate, and multivariate analysis
•
Correlation, or Pearson
r
, or Product-moment coefficient
•
Assumptions of Pearson
r
•
Positive correlation
•
Negative correlation
I.
Review
A.
Univariate, bivariate, or multi-variate analysis implied here?
1.
The average age in your sample is 38, with a standard deviation of 12.82
2.
Chi square test of association between gender and political party affiliation
3.
Effects of group therapy (vs. individual therapy) on increasing score on a
ratio level social support scale
4.
Effects of group therapy (vs. individual therapy) on increasing score on
ratio level social support scale, controlling for gender, ethnicity and SES
B.
Name that statistic:

This
** preview**
has intentionally

**sections.**

*blurred***to view the full version.**

*Sign up*
2
If …
Independent variable is:
And dependent variable is:
Statistical test is:
Categorical (nominal or ordinal)
Categorical (nominal or
ordinal)
__________?
Categorical (nominal—
dichotomous,
two independent
groups)
Ratio or interval
__________?
Categorical (two dependent
groups)
Ratio or interval
__________?
Categorical (nominal or ordinal)
more than two attributes
Ratio or interval
__________?
New!
Continuous
Continuous
Correlation
(Pearson
r
)
II.
What is correlation?
A.
Correlation tests the relationship between a continuous independent variable and a
continuous dependent variable.
B.
Examples of research scenarios:
1.
The number of outpatient therapy sessions utilized is positively correlated
with improvement in depression as measured by the total score of the Beck
Depression Inventory
2.
The number of hours students spend preparing for the 242 exam is
positively correlated with the test score
3.
The number of disruptive behaviors by children in residential care is
negatively correlated with the number of strength-based supportive comments
from staff
C.
Compare to other statistical tests:
1.
Chi-Square: “The variables are associated…”
2.
t
-test: “The group means differ…”
3.
ANOVA: “The group means differ…”
4.
Correlation: “There is a positive [or negative] correlation between the two
variables.”
5.
(Preview to next class) Multivariate regression: “The independent
variables ___, ___, ___, and ___ are related or predictive of the dependent
variable ___, controlling for ____.
D.
Calculation of the Pearson
r